| Free tier | - DeepSeek R1 reasoning model: Addresses complex math, coding, and logical reasoning workflows with chain-of-thought inference support.
- DeepSeek V3 NLP engine: Handles multilingual text generation, summarization, Q&A, translation, and chat interfaces. 
|
Business email with an AI assistant and agent API | Free tier | - **Mailboxes on your own domain**: Create addresses like hello@, support@ or billing@, each with its own login and storage.
- - **Shared team inboxes**: Your whole team can work the same inbox from one dashboard instead of forwarding mail around. Team members who log in are unlimited on Team and Scale plans.
|
Email, calendar and MCP access from your terminal | Freemium | - **Email from the terminal**: Read, search, draft, and send email across supported providers.
- - Calendar management: Work with calendars and events through CLI commands.
|
Email inboxes your AI agents can actually own | Freemium | - **Programmatic inbox creation**: Spin up a dedicated inbox for every agent with one API call. No OAuth flow, no Google Workspace seat, no manual provisioning.
- - **Two-way email, not just sending**: Unlike transactional APIs, AgentMail is built around the inbox. Your agent receives replies, keeps threads intact, and handles attachments and drafts.
|
All-in-one email platform built for SaaS teams | Free tier | - **Simple Email API**: Send transactional emails with a short, readable code snippet. Developers report going from signup to live sending in about 10 minutes.
- - **SMTP relay**: Switch from an existing provider by changing host, port, and credentials. No code rewrite needed if your app already speaks SMTP.
|
Ultra-fast AI inference for real-time apps | Paid | - **LPU inference hardware**: Groq's custom Language Processing Units are built specifically for AI inference, using deterministic execution and on-chip memory to deliver predictable, fast processing.
- - **High-speed inference**: Groq publishes model-specific inference speeds ranging from hundreds to around 1,000 tokens per second on some current models, including 1,000 t/s for GPT-OSS 20B.
|
AI answer engine with real-time, cited web research | Free tier | - **Cited answers in real time**: Every response links back to the sources it pulled from, so you can check the claim instead of trusting it blindly.
- - **Pro Search**: Asks clarifying questions and runs multi-step searches before answering, which produces more precise results on complex queries.
|
AI dubbing and voice cloning in 140+ languages | Freemium | - **Dubbing in 140+ languages**: Upload a video or audio file and get a dubbed version that preserves the original voice, prosody, and emotional delivery instead of flattening it into generic narration.
- - **MARS voice cloning**: Clone a voice from roughly 2 to 3 seconds of reference audio, then use that same voice across every target language you publish in.
|
One API for leading AI models, at lower cost. | Freemium | - **Unified API**: Access GPT, Claude, Gemini, and Grok through a single API instead of maintaining separate provider integrations.
- - **OpenAI-Compatible API**: Continue using compatible OpenAI software development kits by changing the base URL, API key, and model ID.
|
